Tuesday, 2021-03-16

T42<edp_17> As far as I remember I updated all submodules as part of preparation for building 4.0. How recent that change in dhd?00:04
maledp_17 the change was merged a bit over two weeks ago00:19
T42<edp_17> mal: Thanks, that explains. On my other ports I've updated dhd before that.00:21
T42<edp_17> mal: Now I am trying to OTA treltexx 3.3 to 3.4 and got this "Error: nothing provides psmisc-tools needed by patterns-sailfish-device-configuration-treltexx-1+master.20210315161343.c867921-1.2.1.jolla.armv7hl"00:22
T42<edp_17> I guess this is because in pattern I added required: psmisc-tools00:23
T42<edp_17> It seems this is not rquired for 3.4, isn't it?00:23
T42<edp_17> zypper se psmisc doesn't show psmisc-tools.00:24
T42b100dian was added by: b100dian00:38
T42<edp_17> mal: A new error in the i9100 build (local): https://paste.ubuntu.com/p/c6xxys7BTR/00:44
T42<edp_17> It says I need 'pulseaudio-modules-bluez4' but I don't have it. Did I get this error because I sued that pulseaudio-modules-droid-glue repo?00:45
malyou need to use sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper in droid-config-i9100-bluez5 and select correct option01:06
*** Herrie|2 is now known as Herrie01:12
T42<edp_17> mal: I understand that but that port uses bluez4. Do I still need to install bluez5?01:15
malthen install bluez4 of course01:36
malare you sure your patterns are correct?01:36
T42<edp_17> mal: That was the first though for me too. I've double checked all .spec and files and patterns and bluez4 is defined, not bluez5. So, I am unsure where this was coming from.01:38
T42<edp_17> When tried to install droid-config-i9100-bluez4, it also was  complaining about the pulseaudio-modules-bluez4: https://paste.ubuntu.com/p/NwDPsWycZQ/01:43
T42<edp_17> Anyway, I let it as is for now. Thanks for your time and help!01:44
T42<mintphin> I'm having this issue now. Any help? `error: vendor/lineage/build/soong/Android.bp:30:8: module "generated_kernel_includes": cmd: unknown variable '$(PATH_OVERRIDE_SOONG)'`02:10
T42<edp_17> I have OTAd to 3.4 on my treltexx and lipstick doesn't start. It is complaining about "libEGL warning: invalid EGL_PLATFORM given"02:14
T42<edp_17> log: https://paste.ubuntu.com/p/gGrDFx2pqb/02:14
T42<edp_17> droid-hal-device.conf: https://paste.ubuntu.com/p/ZPtpKN6sH3/02:14
T42<lackingcheesepuffs> what are the changes required to make a port from 3.4 to 4.0.1, hybris-16.0?06:58
T42<elros34> here you can find notes about most popular issues/changes: http://collabedit.com/kr9xx07:03
T42<lackingcheesepuffs> Thanks07:04
T42<edp_17> @edp_17 [I have OTAd to 3.4 on my treltexx and lipstick …], I think I know why. Let me check it first.08:06
T42<edp_17> @edp_17 [I have OTAd to 3.4 on my treltexx and lipstick …], Yes, it happend what I thought. Althogh, I added the correct treltexx repo (adaptation-community-treltexx) pointing to release 3.4, but I left the adaptation-community-common repo on the latest devel which was incorrect. Once I amended this latter to 3.4 testing, the updat09:44
T42e was successfull. The device booted up into 3.4 and lipstick started correctly. (Although, BT stopped working.)09:44
T42<edp_17> Now, I am OTA updating the device to 4.0.09:44
T42<edp_17> Many thanks for the help with the OBS!09:44
andreysysHello. Could please anyone to review PR https://github.com/sailfishos/gst-droid/pull/64 ?09:44
T42<edp_17> Hi all, After the OTA to 4.0 I am getting this error: https://paste.ubuntu.com/p/pgX3VBrBWb/09:55
T42<edp_17> Any idea about it? (I have uninstalled patchmanager with all patches and rebooted but the messages didn't go away.)09:56
coderusT42: clear content of /etc/ld.so.preload10:00
T42<edp_17> coderus: How can I do that?10:03
T42<edp_17> coderus: I see. That is a simple ASCII text file. Got it. The issue is gone now. Thanks.10:20
T42<mintphin> I haven't found a solution to my issue yet. An XDA thread told me to resync my repos but that didn't do anything to fix it.12:07
vknechtsrxl: o/ if you're looking to make a "native" port (ie. without android/hybris), the pinephone repo should be a good reference : https://build.merproject.org/project/show/nemo:devel:hw:pine:dontbeevil:pinephone12:07
srxlwonderful, thanks for the tip!12:08
T42<mintphin> Found a fix to my issue.12:34
T42<mintphin> Nevermind.12:37
T42<neochapay> good album hehe12:38
T42<mintphin> Ok I just had to rerun breakfast.12:43
*** vup is now known as __vupbot13:43
*** __vupbot is now known as vup13:44
deathmist_mal: any news on https://github.com/sailfishos-oneplus5/nemo-qml-plugin-systemsettings/commit/377c0a5 still? I've also overridden the 80-udisks2.rules udev rules to avoid creating mount-sd@.service units for the internal partitions: https://github.com/sailfishos-oneplus5/droid-config-cheeseburger/commit/2fe68c114:10
maldeathmist_: there is no PR of that14:15
maldeathmist_: that probably needs some more planning14:22
maldeathmist_: can you show output of the file system layout on your device14:23
T42<adampigg> Its on my todo to work up something for that, but if anyone wants to take it on....!14:23
deathmist_mal: I've catalogged it under https://wiki.postmarketos.org/wiki/OnePlus_5_(oneplus-cheeseburger)#Partition_layout, I hope that help14:34
T42<mintphin> (Photo, 1280x960) https://irc.thaodan.de/.imgstore/zxqffuM0i0.png my build server is locked up14:34
T42<mintphin> (Photo, 1280x720) https://irc.thaodan.de/.imgstore/5GhDuh1uL8.png this is the build server lol14:35
T42<lackingcheesepuffs> @techgeekster [The exact error is: Looking for a fallback sub …], I got this error too14:40
T42<lackingcheesepuffs> @mal [techgeekster: go to hybris/mw/gst-droid and ru …], tried this and it said git is missing14:41
T42<lackingcheesepuffs> how do I install git? zypper in told git is installed14:41
deathmist_@lackingcheesepuffs sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -m sdk-install -R zypper in git, as it's for the cross-compile toolchain14:42
deathmist_mal: you'd likely want some way to discern internal Universal Flash Storage (UFS) from memory cards and USB storage14:44
deathmist_udevadm info dump for my internal sd[a-f] block devices: https://pastebin.com/raw/i8uXUmP914:50
maldeathmist_: how do sd cards appear on that device?14:51
T42<lackingcheesepuffs> @deathmist_ [@lackingcheesepuffs sb2 -t $VENDOR-$DEVICE-$PO …], deathmist_ Should I not run --gg after this?14:51
deathmist_mal: the OnePlus 5 sadly doesn't have any SD card slot, so no idea. maybe @rinigus' device has them?14:51
T42<lackingcheesepuffs> @mal [techgeekster: go to hybris/mw/gst-droid and ru …], after this I mean...14:52
deathmist_@lackingcheesepuffs no, it'll pull upgrades for the packages it builds then and revert your reset to a previous version14:53
T42<elros34> have you checked ATTRS{type}?  SD for sdcard and MMC for internal memory but I am not sure how reliable that is14:53
T42<edp_17> @elros34 : Just wanted to say thanks for the mousetracker rpm. I applied the patch manually and I have the cursore back on my treltexx.👍14:53
rinigusmal & deathmist_ : sd card is on /dev/mmcblk0p1 (partition 1)14:54
rinigusso, it is pretty much reversed to usual case14:54
T42<lackingcheesepuffs> @deathmist_ [@lackingcheesepuffs no, it'll pull upgrades fo …], but then how would I build gmp-droid? I thought —gg would build it but it reported to be missing when running —mic15:09
T42<lackingcheesepuffs> trying with —mw=https://github.com/sailfishos/gmp-droid.git15:11
malbuild_packages.sh --gg checks if gmp-droid is enabled in patterns and only builds it in that case15:15
T42<adampigg> Pro1 has similar patch15:31
T42<elros34> why don't you try ATTRS{type} (/sys/block/*/device/type) property? It should be more reliable than assuming sd card is mmcblk115:34
rinigus@adampigg and deathmist_ : I suspect we will see the changes regarding storage only after Jolla's official devices will experience it. or some of their clients...16:21
malrinigus: is't strange how complicated layout those devices have16:24
malrinigus: can you show udevadm info for sdcard on your device with such an odd layout16:25
rinigusmal: hard to say, they all are from SD8XX line, so they share that. but I think we look for some mechanism (regex) which would allow us to identify SD and local storage. udev does provide such mechanism, as far as I remember, but it is not used in SFOS16:26
rinigusmal: cal you suggest udevadm command?16:26
malrinigus: udevadm info -q all /dev/some_dev16:30
rinigusmal: https://dpaste.org/rk08 that one has /dev/sda and /dev/mmcblk016:32
rinigusmal: using udev rules at https://github.com/sailfishos-sony-tama/droid-config-sony-tama-pie/blob/master/sparse/etc/udev/rules.d/80-udisks2.rules16:34
malrinigus: do all of the sdX have platform-1d8400016:37
malrinigus: or more like PLATFORM_DEVICE=1d84000.ufshc16:37
riniguslet me check16:38
malrinigus: if so then we might be able to use the androidboot.bootdevice for some magic16:38
rinigusmal: yes, they do. (sda, sdb, sdc)16:39
malrinigus: somehow with androidboot.bootdevice we could then make it ignore the internal boot device and mark other things as sd card16:40
rinigusmal: yes, that makes sense16:41
T42<edp_17> Hi all. Can somebody tell me how I can build audiosystem-passthrough-dummy-af on OBS, please? I have defined it in the _pattern but once all build is completed it is not available in the repo.16:42
mal@edp_17 was your build armv7hl based?16:43
*** jusa__ is now known as jusa16:45
T42<edp_17> mal: I think so. Don't remember I built anything else.17:56
T42<elros34> @edp_17 you don't need to build it, it's already buillt in common repo18:07
T42<edp_17> @elros34 : oh, so do I need to remove it from pattern then?18:09
T42<elros34> _pattern?18:09
T42<elros34> you don't even need patterns at obs if you use meta packages. How is that related to your question about buildingsomething at obs?18:11
T42<elros34> I am not sure I follow18:13
T42<edp_17> Yes, I meant _pattern.19:11
T42<elros34> you don't need if you use meta packages unless you want to keep compability with some old releases. Put Requires: audiosystem-passthrough-dummy-af to your patterns/*.inc and rebuild droid-config19:18
T42<mintphin> ok so i'm building the relevant bits of my android base and it's been stuck at `out/soong/.bootstrap/bin/soong_build out/soong/build.ninja` for 1 hour and a half. is that normal?19:46
T42<mintphin> the build machine is clearly not forzen but it is taking a long time19:47
T42<mintphin> *frozen19:49
T42<mintphin> (Video, 7s) https://irc.thaodan.de/.imgstore/kO6VzSVrbw.mp419:49
T42<mintphin> so uh what should i do?19:55
malthat doesn't look normal20:00
malmaybe ctrl+c and try again20:01
T42<mintphin> ok20:01
T42<mintphin> this has happened before and sometimes the laptop i'm building this in freezes20:01
T42<mintphin> i added more swap to mitigate this20:01
T42<edp_17> @elros34 [you don't need if you use meta packages unless …], This is what I've done. That's why I don't understand why I got that error. I'll try again.20:02
T42<elros34> @edp_17 which repo?20:14
T42<mintphin> mal: same issue, fans have stopped spinning, little hard drive activity and still "building"20:29
T42<edp_17> @elros34 : hammerhead: https://build.sailfishos.org/project/show/home:edp17:hammerhead20:34
T42<elros34> so it has correct requires: https://build.sailfishos.org/package/binary/home:edp17:hammerhead/droid-config-hammerhead?arch=armv8el&filename=patterns-sailfish-device-adaptation-hammerhead-1%2Bmaster.20210315230139.6f7d0c0-1.4.1.jolla.armv7hl.rpm&repository=sailfishos_4.0.1.4820:35
T42<edp_17> The 'Requires: audiosystem-passthrough-dummy-af' is in there: https://github.com/edp17/droid-config-hammerhead/blob/master/patterns/patterns-sailfish-device-adaptation-hammerhead.inc#L4920:35
T42<elros34> so make sure you have latest  patterns-sailfish-device-adaptation-hammerhead installed20:35
T42<edp_17> @elros34 [so it has correct requires: https://build.sail …], Yes.20:35
T42<edp_17> On the device?20:36
T42<elros34> sure, this meta package is installed only on device20:36
T42<edp_17> Okay, I'll do that but I though didn't need to install any packages for OTA.20:37
T42<elros34> you have it already installed. I am only saying: make sure you have LATEST version of that packages on device20:38
T42<edp_17> Okay, will do that. Thanks.20:40
T42<mintphin> left it running for more time and it's still showing the same issue21:09
T42<mintphin> at this point i might install ubuntu on a separate ssd and install it on my tower pc21:09
T42<mintphin> *build21:09
T42<elros34> @mintphin just to be sure, did you apply hybris patches before running make?21:11
T42<mintphin> no. i'm on step 5.421:12
T42<mintphin> "build relevant bits of your android base"21:13
T42<mintphin> *building21:13
malwhich android base?21:14
T42<elros34> yeah unfortunately reading hadk pdf is not enough to start. You need to apply hybris patches before building android part: https://github.com/mer-hybris/hadk-faq/#hybris-16-021:15
T42<mintphin> android 1021:15
T42<mintphin> i applied them and i have the same issue21:37
T42<elros34> did you remove out/ directory also21:38
T42<elros34> ?21:38
T42<mintphin> oh no i didn't21:38
T42<mintphin> i'll do that21:38
T42<mintphin> ugh same issue21:56
*** mal__ is now known as mal22:42

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!